Table 4:
Factors associated with household cumulative infection risk (HCIR)a in a rural and an urban community, South Africa, 2020–2021
| HCIR | Univariate | Multivariable | ||
|---|---|---|---|---|
| Variable | n/N (%) | ORe (95% CI) | aORe (95% CI) | |
| Characteristics of the index case | ||||
| Age group (years) | <5 | 18/54 (33) | 7.5 (2.0–27.9) | 5.2 (0.3–21.7) |
| 5–12 | 47/187 (25) | 3.0 (1.1–8.9) | 1.8 (0.6–6.1) | |
| 13–18 | 50/182 (27) | 2.5 (0.9–7.5) | 1.8 (0.6–5.8) | |
| 19–39 | 48/247 (19) | 2.4 (0.8–6.9) | 2.2 (0.7–7.1) | |
| 40–59 | 42/117 (36) | 6.5 (2.0–20.9) | 4.6 (1.3–16.3) | |
| ≥60 | 8/69 (12) | Reference | Reference | |
| Sex | Female | 141/505 (28) | 2.0 (1.2–3.3) | 2.1 (1.2–3.6) |
| Male | 72/351 (21) | Reference | Reference | |
| HIV and viral load copies/mlc | Uninfected | 169/683 (25) | Reference | |
| Infected <400 | 32/107 (30) | 1.1 (0.6–2.2) | ||
| Infected ≥400 | 5/27 (19) | 0.7 (0.2–2.8) | ||
| HIV and/or viral load unknown | 7/39 (18) | 0.9 (0.3–3.2) | ||
| HIV and CD4+ T cell count/mlc | Uninfected | 169/683 (25) | Reference | |
| Infected ≥200 | 31/115 (27) | 1.0 (0.5–1.9) | ||
| Infected <200 | 5/15 (33) | 1.2 (0.2–6.3) | ||
| HIV and/or CD4+ unknown | 8/43 (19) | 0.9 (0.3–3.0) | ||
| BMIb | Underweight | 19/66 (29) | 1.0 (0.4–2.4) | |
| Normal weight | 101/431 (23) | Reference | ||
| Overweight | 38/173 (22) | 0.9 (0.5–1.8) | ||
| Obese | 55/186 (30) | 1.9 (1.1–3.4) | ||
| Symptoms | Absent | 195/775 (25) | Reference | |
| Present | 21/110 (19) | 0.9 (0.4–1.7) | ||
| Duration of viral RNA shedding (days) | ≤4 | 16/171 (9) | Reference | |
| >4 | 200/714 (28) | 4.3 (2.2–8.6) | ||
| Minimum Ct value | ≤30 | 200/720 (28) | 3.6 (1.7–7.6) | 3.9 (1.7–8.8) |
| >30 | 16/146 (11) | Reference | Reference | |
| Epidemic wave | 1 | 17/162 (10) | Reference | |
| 2 | 74/350 (21) | 2.4 (1.1–5.2) | ||
| 3 | 125/373 (34) | 8.9 (4.0–19.9) | ||
| Variant | Wild type | 16/153 (10) | Reference | Reference |
| Beta | 81/366 (22) | 14.6 (0.6–352.3) | 4.2 (1.7–10.1) | |
| Alpha | 4/7 (57) | 2.7 (1.2–6.0) | 14.9 (0.7–324.7) | |
| Delta | 115/359 (32) | 9.5 (4.1–21.8) | 14.6 (5.7–37.5) | |
| Site | Rural | 106/517 (21) | Reference | Reference |
| Urban | 107/339 (32) | 1.6 (0.9–2.9) | 2.0 (1.1–3.9) | |
| Characteristics of the household contact | ||||
| Age group (years) | <5 | 26/112 (23) | 1.4 (0.7–2.8) | 1.3 (0.6–2.7) |
| 5–12 | 73/277 (26) | 1.9 (1.1–3.5) | 2.0 (1.1–3.8) | |
| 13–18 | 39/118 (33) | 2.8 (1.4–5.5) | 3.2 (1.5–6.8) | |
| 19–39 | 31/170 (18) | Reference | Reference | |
| 40–59 | 25/97 (26) | 1.7 (0.8–3.5) | 2.1 (0.9–4.7) | |
| ≥60 | 19/82 (23) | 1.4 (0.6–3.1) | 1.6 (0.7–3.8) | |
| Sex | Female | 130/503 (26) | 1.3 (0.9–1.9) | |
| Male | 83/353 (24) | Reference | ||
| HIV and viral load copies/mld | Uninfected | 179/711 (25) | Reference | |
| Infected <400 | 17/83 (20) | 0.8 (0.4–1.5) | ||
| Infected ≥400 | 9/22 (41) | 1.9 (0.6–5.9) | ||
| HIV and/or viral load unknown | 8/40 (20) | 0.6 (0.2–1.6) | ||
| HIV and CD4+ T cell count/mld | Uninfected | 179/711 (25) | Reference | |
| Infected ≥200 | 25/97 (26) | 1.0 (0.5–1.8) | ||
| Infected <200 | 1/8 (13) | 0.6 (0.1–6.5) | ||
| HIV and/or CD4+ unknown | 8/40 (20) | 0.6 (0.2–1.6) | ||
| Other underlying illness | Absent | 194/765 (25) | Reference | |
| Present | 19/91 (21) | 0.7 (0.4–1.3) | ||
CI - confidence interval, HIV – Human immunodeficiency virus, BMI – Body mass index, Ct – Cycle threshold, rRT-PCR – real-time reverse transcription polymerase chain reaction
Additional factors evaluated but not found to be statistically significant include underlying tuberculosis of index or household contact, BMI of contact, alcohol or smoking of index or contact, SARS-CoV-2 vaccination status of index or contact, fuel used for cooking, water source for handwashing, number of people in household and crowding.
HCIR was defined as the probability of secondary infections within a household following SARS-CoV-2 introduction and estimated by dividing the number of subsequent individuals with confirmed infection within a household cluster following SARS-CoV-2 introduction by the number of susceptible (no previous infection on rRT-PCR or serology) household members. An rRT-PCR-confirmed household cluster was composed of all rRT-PCR-confirmed infection episodes within a household within an interval between the rRT-PCR positive tests of any infection episode pairs of ≤14 days.
BMI=body mass index calculated using the formula (weight in kilograms)/(height in metres squared). We defined BMI categories as follows: underweight - age <18 years weight for age or BMI <−2 standard deviations of the World Health Organization (WHO) Child Growth Standards, age ≥18 years BMI <18.5kg/m2; overweight - age <18 years BMI >+1 and ≤+2 standard deviations of the WHO growth standards, age ≥18 years BMI ≥25 and <30kg/m2, obese – age <18 years BMI >+2 standard deviations of the WHO growth standards, age ≥18 years BMI ≥30 kg/m2
HIV data available for 1401 of 1730 (81%) of index case individuals eligible for this analysis. Among 226 index case PLHIV eligible for this analysis, 224 (99%) had available data on HIV viral load and 219 (97%) had available data on CD4+ T cell count
HIV data available for 1670 of 1730 (97%) of household contact individuals eligible for this analysis. Among 228 household contact PLHIV eligible for this analysis, 217 (95%) had available data on HIV viral load and 214 (94%) had available data on CD4+ T cell count
Estimated using logistic regression adjusted for clustering by site and household